PIN1, cholesterol biosynthesis, and clinical outcomes in urothelial carcinoma (UC): A propensity-weighted real-world clinicogenomic analysis.

Abstract

e16575 Background: PIN1 is a phosphorylation-dependent prolyl isomerase that regulates oncogenic pathways, including cancer stemness, epithelial–mesenchymal transition, immune modulation, and cholesterol biosynthesis. While mechanistic studies implicate PIN1 in aggressiveness across cancer types, its clinical relevance in UC remains unclear. Methods: The Tempus Lens platform was used to identify a real-world cohort of patients (pts) with UC that underwent DNA (Tempus xT) and RNA (Tempus xR) sequencing (N = 4,886). Short variants and copy number changes are reported in our analysis. RNA data were quantified as transcripts per million (TPM) and reported as log2(TPM+1). Pts were classified as PIN1 high (PIN1-h, N = 1,157) or low (PIN1-l, N = 1,211) based on top and bottom mRNA expression quartiles. Single-sample gene set enrichment analysis (ssGSEA) was used to characterize enrichment in MSigDB Hallmark and cholesterol related metabolic pathways. Real-world progression-free survival (rwPFS) and overall survival (rwOS) were measured from initiation of first line (1L) therapy and sample collection date respectively. Propensity score weighting was used for balancing hyperlipidemia history and biopsy site. Hazard ratios (HR) were calculated using multivariable Cox regression. Median rwOS estimated using Kaplan–Meier curves and compared using log-rank tests. Results: The cohort was composed of pts with bladder (83%) and upper tract (17.2%) UC. Of pts with documented staging, 62.9% were metastatic and 16% had a history of hyperlipidemia. PIN1-h tumors were enriched in pathways suggesting increased lipid biosynthesis such as cholesterol regulation ( q < 0.001), glycerolipid metabolism (q < 0.001) and SREBF signaling (q < 0.004) as well as in oncogenic pathways (WNT/Beta Catenin, Notch, p53, KRAS). Immune-signatures (IL6, JAK STAT3 , IFNa, IFNγ) showed decreased enrichment in PIN1-h. Somatic alterations differed, with PIN1-h tumors having higher prevalence of tumor suppressor gene alterations TP53 (68% vs 54%) and RB1 (23% vs 16%) and lower rates of alterations in FGFR3 (11% vs 18%) and ARID1A (18% vs 27%), consistent with aggressive disease biology. rwPFS was shorter in pts with PIN1-h vs PIN1-l tumors when treated with 1L chemotherapy (HR 3.01, 95% CI 1.43-6.33; p = 0.004) but not 1L immunotherapy. Paradoxically, pts with PIN1-h tumors showed a modest but statistically significant improvement in rwOS (HR 0.85, 95% CI 0.74–0.98; p = 0.030). Median rwOS was longer for PIN1-h vs PIN1-l (12.1 vs 15.5 months, p = 0.026). Conclusions: PIN1 expression identifies a distinct subset of UC characterized by cholesterol pathway enrichment and aggressive genomic features. While PIN1-h tumors show shorter rwPFS on 1L chemotherapy, they exhibit longer rwOS overall. These opposing outcomes highlight PIN1 as a complex, potentially treatment-specific prognostic biomarker.